The objective of this study was to define further the environmental requirements for safe sites for germination of diffuse knapweed achenes (seeds). Germination temperature profiles were developed for diffuse knapweed seeds collected from sites in the Great Basin and Colorado. Each profile consisted of seeds germinated at 55 constant or alternating temperatures from 0 through 40 C. The resulting germination was used to develop quadratic response surfaces with regression analysis. Some germination occurred from 71 to 96% of the temperature regimes, depending on the accession being tested. Maximum observed germination ranged from 85 to 98%. Optimum germination, defined as the maximum observed minus one half the confidence interval at the 0.01 level of probability, occurred at a wide range of temperatures from cold periods of 0 through 20 C, alternating with warm periods of 10 through 35 C. The temperature regimes that most frequently supported optimum germination were 5/25 C (5 C for 16 h and 25 C for 8 h in each 24-h period) and 10/25 C. Germination of diffuse knapweed seeds was generally higher at alternating than constant temperatures.

Nomenclature: Diffuse knapweed, Centaurea diffusa Lam. CENDI.

Weedy rice is a serious weed of cultivated rice in most of the rice-growing areas of the tropics and yet, despite its importance as a weed, limited information is available on its biology in Asia. A study was conducted to determine the seed characteristics and growth response to competition of weedy rice variants from five Asian countries (India [IWR], Malaysia [MWR], Philippines [PWR], Thailand [TWR], and Vietnam [VWR]) when grown alone or with either 4 or 12 cultivated rice (‘IR72’) plants. Seed characteristics including weight, length, width, and presence and length of awn differed between weedy rice variants. PWR had the heaviest seed (25 mg seed⁻¹). Growth of weedy rice plants differed among variants and, generally, VWR grew the fastest. IR72 interference greatly reduced tiller and leaf numbers, leaf area, and shoot biomass of all weedy rice variants. At 8 wk after sowing, for example, shoot biomass of weedy rice variants in competition with 12 IR72 plants was 13 to 30% of that where plants were grown alone. TWR plants were shorter than those of other variants at all levels of crop interference. These results show that there are considerable differences among Asian weedy rice variants and yet, competition from the rice crop greatly suppresses weedy rice growth. This suggests that measures to ensure a competitive crop should be an important component of control strategies.

Nomenclature: Weedy rice, Oryza sativa f. spontanea L. ORYSA; rice, Oryza sativa L. ‘IR72’.

Soybean cyst nematode (SCN) is one of the most yield limiting pathogens in U.S. soybean production. Henbit and purple deadnettle are winter annual weeds shown to facilitate SCN reproduction after crop harvest in the eastern Corn Belt. These weeds, along with volunteer soybean that germinates in autumn after harvest, are common to postharvest soybean production fields and provide an opportunity for SCN reproduction and population increase outside of the typical soybean production season. The objective of this experiment was to determine if autumn removal of these weeds and volunteer soybean can influence the winter weed seedbank, plant biomass, and SCN population densities. Microplots were established with or without Lamium spp. and volunteer soybean, and four winter weed removal timings (none, October, December, and May). Dry weights of autumn Lamium spp. were reduced 50% in October when grown in competition with volunteer soybean. SCN juveniles were found in henbit roots at higher densities in October (42 per gram of root) than December (5 per gram of root) and were also found in the roots of volunteer soybean (14 per gram of root) in October. SCN egg population densities were 50% lower in August after the summer fallow period. The results of this experiment suggest that autumn removal of winter annual weeds and volunteer soybean did not reduce SCN populations.

Nomenclature: Henbit, Lamium amplexicaule L. LAMAM; purple deadnettle, Lamium purpureum L. LAMPU; soybean, Glycine max (L.) Merr.; soybean cyst nematode, Heterodera glycines Ichinohe.

Invasive species such as Canada thistle pose a significant threat to ecosystems. The risk of introducing invasive species has increased with human activities, and the effects of such events have economic and aesthetic impacts. Native to Europe, Canada thistle is now established throughout temperate North America. Although there is documentation of early occurrences to North America, little is known on how it has become established in diverse habitats or how it continues to spread. We examined genetic diversity within and among nearly 1,700 Canada thistle individuals from 85 North American locations with the use of seven microsatellite markers in order to address these questions. PAUP and STRUCTURE programs were used to assess genetic diversity and relationships within and between populations. Populations exhibited greater within-population diversity (> 60%) than expected for a reported clonally reproducing species. Total diversity of sampled locations in North America (0.183) was less than previously reported for European locations (0.715), but the greater mean difference between North American populations (0.264 relative to 0.246 from England) suggests strong founder effects or restriction of gene flow influencing individual populations. Furthermore, analyses identified numerous instances where individuals from geographically distant regions clustered together, indicating long-distance translocation of propogules. However, isolation by distance analysis showed significant correlation between location and population genetic distances (r  =  0.1917, P  =  0.006). Within populations, nearly 92% of individuals sampled harbored unique multilocus genotypes, strongly suggesting that sexual reproduction is common. Within populations, analysis of genetic structure indicated significant admixture of genotypes throughout the invasive range in North America. The recurrent distribution of seed throughout North America has led to a highly diverse gene pool and increased the adaptive success Canada thistle to a wide variety of habitats. Future technologies developed for control of Canada thistle should consider this diversity.

Nomenclature: Canada thistle, Cirsium arvense (L.) Scop. CIRAR.

Imidazolinone-resistant (IR) winter wheat allows selective control of jointed goatgrass with the herbicide imazamox. However, the spontaneous hybridization between jointed goatgrass and IR winter wheat threatens the value of the IR technology. The objectives of this study were to determine if F₁ hybrids collected in a commercial production field under IR winter wheat–fallow rotation in Oregon and their first-backcross progeny (BC₁) carried the Imi1 gene and were resistant to imazamox, and to analyze the parentage of F₁ and BC₁ plants. The average seed set of the F₁ spikes was 3.3%, and the average germination of BC₁ seed was 52%. All F₁ and BC₁ plants tested carried Imi1. Jointed goatgrass plant mortality was 100% when treated with imazamox at 0.053 kg ai ha⁻¹, compared to 0% for IR winter wheat and BC₁ progeny. All F₁ plants had jointed goatgrass as the maternal parent; whereas, most BC₁ plants (85.7%) were produced with IR winter wheat as the paternal backcross parent. Although the backcrossing of F₁ hybrids with jointed goatgrass is very low, it demonstrates the potential for introgression of Imi1 from IR winter wheat into jointed goatgrass under natural field conditions.

Nomenclature: Imazamox; jointed goatgrass, Aegilops cylindrica Host., AEGCY; winter wheat, Triticum aestivum L.

Field experiments were conducted from 2004 through 2006 at Pendleton and Clemson, SC, to determine the influence of seasonal emergence of wild radish on phenological development, survival, and seed and biomass production in a noncompetitive environment. The duration of four developmental phases, emergence to bolting, bolting to flowering, flowering to silique production, and silique production to maturity, were recorded following wild radish sowing at monthly intervals from October 2004 through September 2006. Seedling emergence occurred 2 to 4 wk after sowing. Mortality of seedlings that emerged from December through March was greater than that of seedlings that emerged in all other months. Wild radish that emerged from April through August completed its life cycle by summer or early autumn. Wild radish that emerged from September through November was able to survive the winter and complete its life cycle the following spring. The developmental phases most affected by time of emergence were emergence to bolting and bolting to flowering. The duration of emergence to bolting ranged from 249 to 479 growing degree days (GDD), and bolting to flowering from 270 to 373 GDD, depending on the month of emergence. The total life cycle of wild radish varied from a low of 1,267 GDD following June emergence to 1,503 GDD following November emergence. Multiple regression analysis revealed that emergence to bolting and silique production to maturity phases were dependent on accumulated heat units and photoperiod. Seed and biomass production were influenced by month of emergence. An average of 1,470 seeds plant⁻¹ was produced when emergence occurred in July and 10,170 seeds plant⁻¹ when emergence occurred in November. Plants that emerged in autumn exhibited minimal growth during the winter months, but conditions were conducive for growth in mid-March and April, with biomass production of 809 g plant⁻¹ at silique production.

Nomenclature: Wild radish, Raphanus raphanistrum L. RAPRA.

The spread and dominance of the invasive native hay-scented fern in the understory is one of the most significant changes that has affected the forest ecosystems in the northeastern United States in the last century. We studied changes in the distribution and dynamics of hay-scented fern at a large scale over a 10-yr period in Pennsylvania. The study included 56 stands covering 1,009 ha in two ecoregions. Hay-scented fern was more widely distributed and occurred at higher densities in the Allegheny Plateau ecoregion vs. the Ridge and Valley. Hay-scented fern abundance was positively associated with overstory red maple abundance in both ecoregions. After overstory removal, the density and distribution of hay-scented fern tended to increase and remain at elevated levels in stands that were not treated with herbicide. Herbicide treatments resulted in temporary reductions in fern densities and created a “window of opportunity” for the establishment of tree regeneration.

Nomenclature: Hay-scented fern, Dennstaedtia punctilobula (Michx.) Moore; red maple, Acer rubrum L.

Horseweed can be a problematic weed in no-till soybean fields and populations can vary in their response to 2,4-D. The objective of this study was to evaluate the growth and seed production of four horseweed populations after exposure to 2,4-D. 2,4-D amine was applied at 0, 140, 280, and 560 g ae ha⁻¹ to 5- to 10-cm-tall horseweed plants. An additional treatment of 280 g ha⁻¹ of 2,4-D 840 g ae ha⁻¹ of glyphosate was included in the study. At 2 wk after treatment (WAT), injury ranged from 47 to 98%, but by 6 WAT the injury ranged from 89 to 100% for all four populations. Between 6 and 12 WAT some individual horseweed plants started to recover. No differences in dry weights were observed between the four populations in the untreated checks at 0, 2, 6, and 12 WAT. At 280 g ha⁻¹ of 2,4-D amine, seed production was reduced by greater than 95%. However, three of the four horseweed populations had plants that survived and produced seed after exposure to 840 g ha⁻¹ of glyphosate 280 g ha⁻¹ of 2,4-D. One plant produced seed after exposure to 560 g ha⁻¹ of 2,4-D. These results suggest that horseweed can evolve resistance to 2,4-D and no fitness penalities were observed in populations that had higher levels of tolerance to 2,4-D.

Nomenclature: 2,4-D amine; horseweed, Conyza canadensis L. (Cronq.), ERICA; soybean, Glycine max (L.) Merr.

This study examined seed ultrastructure in relation to germination of North American dandelion seeds. Based on laboratory rearing observations, it was thought that the design of the pappus acts as a conduit facilitating water entry into the seed. It was hypothesized that seeds without a pappus would yield fewer seedlings and require more time to germinate than seeds with an intact pappus. Seed ultrastructure was investigated using scanning electron microscopy, while relative humidity and fungal association were explored as factors that may confer an advantage to intact seeds. Results indicate that germination for seeds lacking a pappus is 31% lower than control seeds (with an intact pappus) and that the seeds lacking a pappus require more time to germinate. Relative humidity did not differentially affect germination, and while a fungus Cladosporium cladosporioides was recovered internally, its presence neither enhanced germination nor decreased time to germination when tested by antimycotic removal. Electron micrographs revealed that (1) the pappus is hollow and (2) the pericarp of the fruit fuses with and partially encloses the pappus. Fusion of the pappus with the fruit suggests that this structure acts as a device to regulate seed hydration.

Nomenclature: North American dandelion, Taraxacum officinale G. H. Weber ex Wiggers; Cladosporium cladosporioides.

Experiments were conducted on Palmer amaranth seeds collected in 2004 and 2006 from a natural population near Pendleton, SC, to determine the temperature and light requirements for germination of seeds retrieved from soil surface or from 10-cm depth in the field. A cyclic change in seed germination of Palmer amaranth in response to temperature and light occurred during a 12-mo after-ripening period. Freshly matured seeds collected in November required mean temperatures ≥ 25 C, and natural or red (R) light for increased germination. Following after-ripening in winter, seeds experienced a reduction in dormancy and germinated higher at 25 to 35 C mean compared with 10 to 15 C mean. With after-ripening for an additional 3 mo in May, seeds experienced a broadening of thermal range (10 to 40 C mean), and germination in natural light or R light was more than twice the germination in the absence of light. Fluctuating temperatures (7.5 C amplitude) improved germination over constant temperatures, except in summer and fall (9 and 12 mo after seed maturation). Exposure of seeds to high temperatures during summer caused secondary dormancy induction. Averaged over thermal amplitudes, seeds retrieved in fall required mean temperatures > 25 C for increased germination. Burial in spring for 3 to 6 mo induced seed dormancy, and the relative germination in fall (12 mo after seed maturation) was at least 50% higher for seeds retrieved from soil surface compared to seeds exhumed from 10-cm soil depth. Seeds retrieved in late summer and fall required natural light or R light for promoting germination, whereas far-red (FR) light or darkness inhibited germination. Furthermore, the effect of R and FR light was reversible, indicating a partially phytochrome-mediated germination response of Palmer amaranth seeds following 9 to 12 mo of after-ripening in the field.

Nomenclature: Palmer amaranth, Amaranthus palmeri S. Wats. AMAPA.

Field pea seed from bin cleaning operations stored overwinter on nearby cropland was observed to correlate with weed and crop growth suppression for up to three subsequent years. To explore the phenomenon more explicitly, plant growth suppression trials were undertaken with soil sampled 18 mo apart from two locations that had contained field pea seed residues. Test plant species grown in the residue-affected and nearby residue-free soils were compared in greenhouse experiments. Germination was either fully inhibited or emergence was delayed by more than one week. Dry matter accumulation of test species grown in residue-affected soil was significantly reduced compared to dry matter of these test species grown in residue-free soil (P < 0.0001). Canola and field pea were inhibited more than wheat and green foxtail over both years. Greenhouse trials also revealed that germination of wild oat was inhibited in the residue-affected soils, although wheat and grassy weeds were less suppressed than dicots overall. Significant reductions of weed species diversity and abundance were correlated to residue-affected soils (P < 0.0001) when compared to residue-free soils using multi-response permutations procedures. Germination of wheat and canola seed was inhibited, using aqueous extracts of weathered pea seeds or extracts of the residue-affected soil in bioassays in sterile media. An allelopathic response was proposed to explain the above results, indicating a need for further research on this system. Weed management strategies could be developed with field pea seed residues to provide innovative weed control techniques.

Nomenclature: Green foxtail, Setaria viridis L. Beauv. SETVI; wild oats, Avena fatua L. AVEFA; canola, Brassica napus L; field pea, Pisum sativum L; wheat, Triticum aestivum L.

Planting peanut in narrow rows for weed control has not been investigated in recently released Virginia market peanut cultivars. Research was conducted in North Carolina from 2007 to 2009 to determine the effect of cultivar, planting pattern, and level of weed management inputs on weed control, peanut yield, and estimated economic return. Experiments consisted of three levels of weed management (clethodim applied POST, cultivation and hand-removal of weeds, and clethodim and appropriate broadleaf herbicides applied POST), three levels of planting pattern (single rows spaced 91 cm apart, standard twin rows spaced 20 cm apart on 91-cm centers, and narrow twin rows consisting of twin rows spaced 20 cm apart on 46-cm centers), and two Virginia cultivars (‘NC 12C’ and ‘VA 98R’). Weed management affected common lambsquarters, common ragweed, eclipta, nodding spurge, pitted morningglory, Texas millet, and yellow nutsedge control, irrespective of cultivar or planting pattern. Cultivar and planting pattern had only minor effects on weed control and interactions of these treatment factors seldom occurred. Weed control achieved with cultivation plus hand-removal was similar to weed management observed with grass and broadleaf herbicide programs. Pod yield did not differ among treatments when broadleaf weeds were the dominant species but did differ when Texas millet was the most prevalent weed. The highest yield with conventional herbicide weed management was in standard twin and narrow twin row planting patterns, although no differences among planting patterns were noted when cultivation and hand-removal were the primary weed management tactics. Differences in estimated economic return were associated with weed species, and interactions of treatment factors varied by year for that parameter.

Nomenclature: Clethodim; common lambsquarters, Chenopodium album L. CHEAL; common ragweed, Ambrosia artemisiifolia L. AMBEL; eclipta, Eclipta prostrata (L.) L. ECLAL; nodding spurge, Euphorbia nutans Lag. EPHNU; pitted morningglory, Ipomoea lacunosa L. IPOLA; Texas millet, Urochloa texana (Buckl.) R. Webster. PANTE; yellow nutsedge, Cyperus esculentus L. CYPES; peanut, Arachis hypogaea L. ‘NC 12C’ and ‘VA 98R’.

Certain sulfonylurea (SU) herbicides are used to remove overseeded cool-season species from bermudagrass. The effects of nitrogen (N) on the efficacy of a new SU herbicide, flazasulfuron, have not been determined. Field and laboratory studies were conducted in 2008 and 2009 evaluating the efficacy of flazasulfuron for control of overseeded perennial ryegrass contaminated with annual bluegrass. Flazasulfuron was applied at rates of 4.4, 8.8, and 17.5 g ha⁻¹ alone, and in between sequential applications of N fertilizer at 73 kg N ha⁻¹. N was granularly applied immediately prior to herbicide treatment and 4 wk later. In both years, the level of annual bluegrass control with flazasulfuron and two applications of N at 73 kg N ha⁻¹ was significantly greater than following treatment with flazasulfuron alone. This response was observed for all application rates of flazasulfuron on every rating date. The level of annual bluegrass control with flazasulfuron at 4.4 g ha⁻¹ and two applications of N at 73 kg ha⁻¹ was greater than flazasulfuron at 17.5 g ha⁻¹ alone each year. No significant differences in perennial ryegrass control were observed for flazasulfuron with and without N fertility. In laboratory studies with annual bluegrass, treatment with N fertilizer at 73 kg N ha⁻¹ increased translocation of ¹⁴C flazasulfuron (and any potential metabolites) from treated annual bluegrass leaves to other shoot tissues by 18% at 1 h after treatment and 22% at 4 h after treatment compared to plants not treated with N fertilizer. This increase in translocation may explain the increased level of annual bluegrass control observed in the field.

Nomenclature: Flazasulfuron; annual bluegrass, Poa annua L. POAN; bermudagrass, Cynodon dactylon (L.) Pers. CYNDA; perennial ryegrass, Lolium perenne L. LOLPE.

Crops that effectively compete with weeds may be more suitable in low-input agricultural systems. A 2-yr field experiment was conducted in northern Greece to assess the competitiveness of seven annual aromatic plants (anise, sweet fennel, sweet basil, dill, coriander, parsley, and lacy phacelia) on common purslane, common lambsquarters, black nightshade, and barnyardgrass. The phytotoxicity of the essential oils produced by these aromatic plants was also determined using a perlite-based bioassay with barnyardgrass. Separation, identification, and quantification of the volatile compounds of these essential oils were also performed. After the harvest of the aromatic plants (8 wk after planting), the greatest weed fresh weight reduction (94 to 100%) was recorded in lacy phacelia, whereas the least (0 to 30%) was recorded in parsley. Lacy phacelia and sweet fennel produced the greatest fresh biomass yield in weedy and weed-free treatments, whereas parsley, dill, and coriander produced the lowest. Biomass of sweet fennel and anise was reduced by only 9 to 11% by weed competition, whereas biomass of lacy phacelia was not significantly affected. The essential oils isolated from sweet fennel and sweet basil were the most phytotoxic on barnyardgrass, whereas those isolated from lacy phacelia and anise were the least phytotoxic. Conclusively, aromatic plants with great competitiveness such as lacy phacelia, anise, and sweet fennel provided great weed suppression and they could be cultivated with low inputs in herbicides. However, high competitiveness of aromatic plants may not always be correlated with high essential oil phytotoxicity.

Nomenclature: Anise, Pimpinella anisum L; barnyardgrass, Echinochloa crus-galli (L.) Beauv. ECHCG; black nightshade, Solanum nigrum L. SOLNI; common lambsquarters, Chenopodium album L. CHEAL; common purslane, Portulaca oleracea L. POROL; coriander, Coriandrum sativum L; dill, Anethum graveolens L; lacy phacelia, Phacelia tanacetifolia Benth; parsley, Petroselinum crispum (P. Mill.) Nyman ex A.W. Hill; sweet basil, Ocimum basilcum L; sweet fennel, Foeniculum vulgare P. Mill.

Preventing the introduction of weeds into the farming system through sowing of clean seeds is an essential component of weed management. The weed seed contamination of cleaned grain and herbicide resistance levels of the recovered weed seeds were examined in a study conducted across 74 farms in the Western Australian grainbelt. Most farmers grew and conserved their own crop seed. The majority of cleaned samples had some level of seed contamination from 11 foreign weed and volunteer crop species, with an average of 62 seeds 10 kg⁻¹ grain, substantially higher than the 28 seeds 10 kg⁻¹ grain expected by farmers. The most common weed contaminants across all samples were rigid ryegrass, wild radish, brome, and wild oat. When categorized by crop type, rigid ryegrass was the most frequent contaminant of cereal crops (barley and wheat), however wild radish was the most frequent contaminant of lupin crops. Uncleaned crop seed samples had almost 25 times more contamination than cleaned crop seed. Herbicide resistance was highly prevalent within rigid ryegrass populations recovered from cleaned grain except for glyphosate, which controlled all populations tested. Some resistance was also found in wild radish and wild oat populations; however, brome was susceptible to fluazifop. This study has shown that farmers are unknowingly introducing weed seeds into their farming systems during crop seeding, many of which have herbicide resistance.

Nomenclature: Fluazifop; glyphosate; brome, Bromus spp; rigid ryegrass, Lolium rigidum Gaudin; wild oat, Avena fatua L; wild radish, Raphanus raphanistrum L; barley, Hordeum vulgare L; lupin, Lupinus spp; wheat, Triticum aestivum L.

Laboratory studies were conducted to compare the soil adsorption of aminopyralid and clopyralid with the use of batch-slurry and centrifugation assays. The calculated soil binding constants for both herbicides varied between the two techniques, but the centrifugation assay had a lower coefficient of variation compared to the batch-slurry assay. These results indicate that a centrifugation assay is a more accurate procedure for measuring the interaction of aminopyralid and clopyralid with soils. Aminopyralid adsorbed more tightly than clopyralid to six of the eight soils tested. Adsorption K_d values ranged from 0.083 to 0.364 for clopyralid and 0.106 to 0.697 for aminopyralid. Pearson correlation analysis indicated that binding of both herbicides was highly correlated to soil organic matter and texture but not to soil pH. On average, soil thin-layer chromatography indicated that aminopyralid was less mobile (R_f  =  0.82) than clopyralid (R_f  =  0.91), although both were mobile. These results suggest that aminopyralid will have a lower leaching potential than clopyralid. Lower potential aminopyralid soil leaching, coupled with low use rates, suggests it may be the herbicide of choice in areas where potential for leaching could be a concern.

Nomenclature: Aminopyralid; clopyralid.

Field studies examined the interaction of soil pH with differing levels of atrazine exposure over 4 yr. Soil pH was 5.2 to 7.1 with atrazine exposures ranging from none (0) to eight applications over a 4-yr period (for each year, one application at planting and one early postemergence). The entire plot area was uniformly managed to reduce potential confounding effects due to cropping history, tillage, and other factors. Soil from all plots previously treated with atrazine displayed rapid atrazine dissipation, with half-lives under laboratory conditions being less than 4 d in plots of pH 5.5 or greater and less than 8 d in the field. Soil pH had a marked effect, with slower atrazine dissipation in those plots that had a pH 5.5 or less. This effect of pH and previous atrazine history was consistent in laboratory and field environments. Implications of these findings include probable reduction in weed control due to more rapid atrazine dissipation and potentially reduced loadings into surface water due to this phenomenon.

Nomenclature: Atrazine.

Research was conducted to determine picloram and aminopyralid sorption in five soils and three clay minerals and to determine if the potential for off-target movement of aminopyralid in soil is less than that of picloram. Nearly all sorption of picloram and aminopyralid occurred between 0 and 8 h, and the maximum theoretical sorption of picloram and aminopyralid were 10.3 and 15.2%, respectively. Freundlich distribution coefficients (K_f) for picloram ranged from 0.12 in a Cecil sandy loam to 0.81 in an Arredondo fine sand, while K_f values for aminopyralid ranged from 0.35 in a Cecil sandy loam to 0.96 in an Arredondo fine sand. Furthermore, K_f values of aminopyralid were higher than those of picloram in all soils tested. K_f values of picloram in clay minerals were 0.25 (kaolinite), 1.17 (bentonite), and 1,016.4 (montmorillonite), and those of aminopyralid were 5.63 (kaolinite), 2.29 (bentonite), and 608.90 (montmorillonite). It was concluded that soil sorption of aminopyralid was greater than that of picloram and that the potential for off-target movement of aminopyralid is less than that of picloram.

Nomenclature: Aminopyralid; picloram.

Most studies on weed population dynamics in farming systems have focused on the effects of different weed control strategies. Those studies usually assume that farmers, operating within a particular system, have a uniform management style. However, it is likely that weed management decision making also varies between farmers that operate within a system. In this study, the relationship between weed management behavior and the outcome of that behavior within an organic farming system is studied. It is hypothesized that differences in weed pressure between organic farms can be related to differences in farmers' weed management behavior. We explore which weed and general management factors are of main influence on the weed pressure, and investigate the influence of farmer's beliefs and knowledge on weed control techniques and the observed weed pressure. Preventive measures and timing of main soil tillage operation were identified as the weed management factors most influential for weed pressure. With the increasing number of preventive measures applied, weed pressure decreased, with a stale seedbed being the most important preventive measure. The weed pressure increased with the number of days after September 1st on which the main tillage operation was carried out. Because of this postponement of the tillage treatments, the growing season of weeds was extended and more species were able to reproduce before winter, thereby enhancing weed pressure. Field size, rather than weed pressure, determined the number of hand-weeding hours per ha; with increasing field size the amount of hand weeding per surface area was reduced. On farms with lower weed pressures a higher percentage of competitive crops were grown than on farms with higher weed pressures. The farmer's beliefs and knowledge on weed control techniques differed between farmers with different weed pressures. Market-oriented growers had a higher on-farm weed pressure than crop-growth–oriented growers. It was concluded that studies on weed management behavior and the effect of that behavior can lead to a better understanding of farming systems and to more effective weed management in those systems.

WeedML is a proposed standard to formulate models of weed demography (or perhaps even complex models in general) that are both transparent and straightforward to reuse as building blocks for new models. The paper describes the design and thoughts behind WeedML which relies on XML and object-oriented systems development. Proof-of-concept software is provided as open-source C code and executables that can be downloaded freely.

The human dimension of weed management is most evident when farmers make decisions contrary to science-based recommendations. Why do farmers resist adopting practices that will delay herbicide resistance, or seem to ignore new weed species or biotypes until it is too late? Weed scientists for the most part have ignored such questions or considered them beyond their domain and expertise, continuing to focus instead on fundamental weed science and technology. Recent pressing concerns about widespread failure of herbicide-based weed management and acceptability of emerging technologies necessitates a closer look at farmer decision making and the role of weed scientists in that process. Here we present a circular risk-analysis framework characterized by regular interaction with and input from farmers to inform both research and on-farm risk-management decisions. The framework utilizes mental models to probe the deeply held beliefs of farmers regarding weeds and weed management. A mental model is a complex, often hidden web of perceptions and attitudes that govern how we understand and respond to the world. One's mental model may limit ability to develop new insights and adopt new ways of management, and is best assessed through structured, open-ended interviews that enable the investigator to exhaust the subjects inherent to a particular risk. Our assessment of farmer mental models demonstrated the fundamental attribution error whereby farmers attributed problems with weed management primarily to factors outside of their control, such as uncontrolled weed growth on neighboring properties and environmental factors. Farmers also identified specific processes that contribute to weed problems that were not identified by experts; specifically, the importance of floods and faulty herbicide applications in the spread of weeds. Conventional farmers expressed an overwhelming preference for controlling weeds with herbicides, a preference that was reinforced by their extreme dislike for weeds. These preferences reflect a typical inverse relationship between perceived risk and benefit, where an activity or entity we perceive as beneficial is by default perceived as low risk. This preference diminishes the ability of farmers to appreciate the risks associated with overreliance on herbicides. Likewise, conventional farmers saw great risk and little benefit in preventive measures for weed control. We expect that thorough two-way communication and a deeper understanding of farmer belief systems will facilitate the development of audience-specific outreach programs with an enhanced probability of affecting better weed management decisions.